Speech recognition systems on the cell broadband engine processor

被引：0
|
作者：
Liu, Yang [1 ]
Jones, Holger [1 ]
Vaidya, Sheila [1 ]
Perrone, Michael P. [2 ]
Tydlitát, Bořivoj [3 ]
Nanda, Ashwini K. [2 ]
机构：
[1] Lawrence Livermore National Laboratory, 7000 East Avenue, Livermore, CA 94550, United States
[2] IBM Research Division, Thomas J. Watson Research Center, P.O. Box 218, Yorktown Heights, NY 10598, United States
[3] IBM Czech Republic, Voice Technologies and Systems, V Parku 2294/4, 148 00 Praha 4, Czech Republic
来源：
IBM Journal of Research and Development | 2007年 / 51卷 / 05期
关键词：
In this paper we describe our design; implementation; and initial results of a prototype connected-phoneme-based speech recognition system on the Cell Broadband Engine™ (Cell/B.E.) processor. Automated speech recognition decodes speech samples into plaintext (other representations are possible) and must process samples at real-time rates. Fortunately; the computatioinal tasks involved in this pipeline are highly data parallel and can receive significant hardware acceleration from vector-streaming architectures such as the Cell/B.E. Architecture. Identifying and exploiting these parallelism opportunities is challenging and critical to improving system performance. From our initial performance timings; we observed that a single; Cell/B.E. processor can recognize speech from thousands of simultaneous voice channels in real time - a channel density that is orders of magnitude greater than the capacity of existing software speech recognizers based on CPUs (central processing units). This result emphasizes the potential for Cell/B.E. processor-based speech recognition and will likely lead to the development of production speech systems using Cell/B.E. processor clusters. © Copyright 2007 by International Business Machines Corporation;
D O I：
暂无
中图分类号：
学科分类号：
摘要：
Journal article (JA)
引用
收藏
页码：583 / 591
相关论文
共 50 条
  • [41] Digital signal processor implementation of active noise control systems for broadband noise cancellation in engine exhaust systems
    Wu, Jian-Da
    Bai, Mingsian R.
    Japanese Journal of Applied Physics, Part 1: Regular Papers and Short Notes and Review Papers, 2000, 39 (08): : 4982 - 4986
  • [42] Digital signal processor implementation of active noise control systems for broadband noise cancellation in engine exhaust systems
    Wu, JD
    Bai, MR
    JAPANESE JOURNAL OF APPLIED PHYSICS PART 1-REGULAR PAPERS SHORT NOTES & REVIEW PAPERS, 2000, 39 (08): : 4982 - 4986
  • [43] LINGUISTIC PROCESSOR IN A CONVERSATIONAL SPEECH RECOGNITION SYSTEM
    SHIKANO, K
    KOHDA, M
    REVIEW OF THE ELECTRICAL COMMUNICATIONS LABORATORIES, 1978, 26 (11-1): : 1505 - 1520
  • [44] VERSATILE VECTOR PROCESSOR FOR MULTICHANNEL SPEECH RECOGNITION
    OSBORN, RR
    JOURNAL OF THE ACOUSTICAL SOCIETY OF AMERICA, 1979, 65 : S132 - S132
  • [45] A DYNAMIC-PROGRAMMING PROCESSOR FOR SPEECH RECOGNITION
    QUENOT, GM
    GAUVAIN, JL
    GANGOLF, JJ
    MARIANI, JJ
    IEEE JOURNAL OF SOLID-STATE CIRCUITS, 1989, 24 (02) : 349 - 357
  • [46] ACOUSTIC PROCESSOR IN A CONVERSATIONAL SPEECH RECOGNITION SYSTEM
    NAKATSU, R
    KOHDA, M
    REVIEW OF THE ELECTRICAL COMMUNICATIONS LABORATORIES, 1978, 26 (11-1): : 1486 - 1504
  • [47] A COMPARISON OF 2 TRAINING STRATEGIES FOR SPEECH RECOGNITION WITH AN ELECTROTACTILE SPEECH PROCESSOR
    ALCANTARA, JI
    COWAN, RSC
    BLAMEY, PJ
    CLARK, GM
    JOURNAL OF SPEECH AND HEARING RESEARCH, 1990, 33 (01): : 195 - 204
  • [48] Real-time mutual-information-based linear registration on the Cell Broadband Engine processor
    Ohara, Moriyoshi
    Yeo, Hangu
    Savino, Frank
    Iyengar, Giridharan
    Gong, Leiguang
    Inoue, Hiroshi
    Komatsu, Hideaki
    Sheinin, Yadim
    Daijavad, Shahrokh
    Ericksons, Bradley
    2007 4TH IEEE INTERNATIONAL SYMPOSIUM ON BIOMEDICAL IMAGING : MACRO TO NANO, VOLS 1-3, 2007, : 33 - +
  • [49] Monte Carlo Simulations of Spin Glass Systems on the Cell Broadband Engine
    Belletti, Francesco
    Guidetti, Marco
    Maiorano, Andrea
    Mantovani, Filippo
    Schifano, Sebastiano Fabio
    Tripiccione, Raffaele
    PARALLEL PROCESSING AND APPLIED MATHEMATICS, PT I, 2010, 6067 : 467 - +
  • [50] Recognition of conversational telephone speech using the JANUS speech engine
    Zeppenfeld, T
    Finke, M
    Ries, K
    Westphal, M
    Waibel, A
    1997 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING, VOLS I - V: VOL I: PLENARY, EXPERT SUMMARIES, SPECIAL, AUDIO, UNDERWATER ACOUSTICS, VLSI; VOL II: SPEECH PROCESSING; VOL III: SPEECH PROCESSING, DIGITAL SIGNAL PROCESSING; VOL IV: MULTIDIMENSIONAL SIGNAL PROCESSING, NEURAL NETWORKS - VOL V: STATISTICAL SIGNAL AND ARRAY PROCESSING, APPLICATIONS, 1997, : 1815 - 1818
12345